Tests are tied to implementation too closely

[purplecabbage]

All commands flag tests test for the specific values of the flags.
This means that we essentially have written the same code in the test+implementation, and if we ever want to change the implementation we have to do it in 2 places.

example:

ActionList.flags = {
  ...RuntimeBaseCommand.flags,
  // example usage:  aio runtime:action:list --limit 10 --skip 2
  'limit': flags.integer({
    char: 'l',
    description: 'only return LIMIT number of actions from the collection (default 30)',
    hidden: false, // hide from help
    multiple: false, // allow setting this flag multiple times
    required: false // not mandatory
  }),
  'skip': flags.integer({
    char: 's',
    description: 'exclude the first SKIP number of actions from the result',
    multiple: false, // allow setting this flag multiple times
    required: false // not mandatory
  }),
  'json': flags.boolean({
    description: 'output raw json'
  })
}

and the test:

test('flags', async () => {
  expect(TheCommand.flags).toMatchObject({
    'limit': {
      'char': 'l',
      'description': 'only return LIMIT number of actions from the collection (default 30)',
      'hidden': false,
      'multiple': false,
      'required': false
    },
    'skip': {
      'char': 's',
      'description': 'exclude the first SKIP number of actions from the result',
      'multiple': false,
      'required': false
    }
  })
})

[adobe-bot]

JIRA issue created: https://jira.corp.adobe.com/browse/ACNA-590